Solution for Molecular Modeling and Biomedical Data Analytics Test Task

Overview

This README documents my solution to a test task involving molecular modeling and analysis of biomedical data. The project focused on identifying a promising drug candidate for the treatment of Type 2 Diabetes Mellitus through in silico analysis and experimental validation.

Methodology

• Data Analysis: Utilized Python and various libraries (Pandas, Seaborn, Matplotlib, Biopython) to analyze a dataset of ~100k drug candidates, narrowing down to 10 leads based on in silico parameters.
• Hypothesis Formulation: Developed a scientific hypothesis on the relationship between the predicted cellular activity of the leads against GLP-1R and GIPR receptors and their in silico descriptors. Additional descriptors like aromaticity and instability index were calculated using Biopython.
• Experimental Validation: Collaborated with a team to validate the hypothesis experimentally, assessing the cellular activity of the shortlisted molecules in vitro.
• Molecular Dynamics (MD) Simulation: Prepared, equilibrated, and conducted MD simulations using GROMACS to characterize the most promising molecule's stability and interactions with the target receptors.
• Visualization: Employed PyMOL and py3Dmol for 3D visualization of peptide ligands and their complexes with target receptors, ensuring a thorough structural understanding.

Results

• The experimental results supported the initial hypothesis to some extent, revealing the complex nature of molecular interactions with receptors.
• The MD simulation and analysis indicated the stability of the selected lead compound in complex with the receptor, as evidenced by consistent RMSD values.

Conclusion

The combined in silico and experimental approach provided valuable insights into the molecular basis of the selected lead's interaction with GLP-1R and GIPR receptors. This lead, identified through rigorous analysis, holds potential for further development in treating Type 2 Diabetes Mellitus.

Additional Files

• 3D structures of the ligand and ligand-receptor complex
• Python notebooks and scripts used for data analysis and visualization
• MD simulation input and output files

For a more detailed walkthrough of the analysis, please refer to the Python notebooks included in the repository.